The potential problem with that is that you invite ground current through the shield, which may lead to hum/noise.
In the ideal case, the shield carries no (ground) current and is only there to carry the reference for the signal. That is the great advantage of balanced connections - the ground is carried separately from the signal & reference.
Jan
In the ideal case, the shield carries no (ground) current and is only there to carry the reference for the signal. That is the great advantage of balanced connections - the ground is carried separately from the signal & reference.
Jan
If you keep the shield isolated from chassis at the input connector, a small cap 2-3nF front the connector barrel to the chassis will still give very good RF immunity because at those frequencies and he shield and housing are a single enclosure.
I agree with Jan - ultimately a balanced interconnection is best but it’s not as challenging as unbalanced. Where the fun in that?
I agree with Jan - ultimately a balanced interconnection is best but it’s not as challenging as unbalanced. Where the fun in that?
remind me - PE as in Protective Earth? You want to lift that to correct a design oversight? Are you competing for the Darwin Award? ;-)
Yes. Have you read the posts by ilimzn here? https://www.diyaudio.com/forums/solid-state/324413-dozens-schemes-wire-amp-4.html#post5477189
and here https://www.diyaudio.com/forums/solid-state/324413-dozens-schemes-wire-amp-5.html#post5477333 especially
and here https://www.diyaudio.com/forums/solid-state/324413-dozens-schemes-wire-amp-5.html#post5477333 especially
Last edited:
I think this is the important post if the other link fails for anyone:
The dozens schemes to wire an amp...
I wish someone could at least say which section, of which document contains this information. After all I can only sell one kidney to pay for the standard.
The dozens schemes to wire an amp...
I wish someone could at least say which section, of which document contains this information. After all I can only sell one kidney to pay for the standard.
Whatever, if you are doing DIY and the chassis is conductive or any part can be exposed to mains, it must be earthed.
If anything is going to be 'lifted' it must be the internal electronics (via high current anti-parallel diodes) and not the chassis.
I personally don't think anyone should be building DIY stuff and claiming its double insulated. Big commercial outfits spend a lot of money meeting standards, and certifying equipment both for safety reasons and to make sure if something goes wrong, they can claim due process.
However, I've also seen claims that the input connectors must be earthed directly to the metal work for safety in case the attached equipment is faulty are not correct either. Each piece of gear has to meet safety standards, and no equipment can rely on another piece of equipment to meet safety standards.
If anything is going to be 'lifted' it must be the internal electronics (via high current anti-parallel diodes) and not the chassis.
I personally don't think anyone should be building DIY stuff and claiming its double insulated. Big commercial outfits spend a lot of money meeting standards, and certifying equipment both for safety reasons and to make sure if something goes wrong, they can claim due process.
However, I've also seen claims that the input connectors must be earthed directly to the metal work for safety in case the attached equipment is faulty are not correct either. Each piece of gear has to meet safety standards, and no equipment can rely on another piece of equipment to meet safety standards.
Yes, and no one is saying otherwise. The question is about the potential difference between the equipment and earth
Yes. Seems very difficult and expensive to find the actual documentation. Regards the UK it appears below 50V is considered safe
Balanced connections for short connections, say less than 1 m, is a bit overengineering.
Like small cap front the connector barrel to the chassis the immunity they provide is known to be limited at very high frequencies. Recently a member reported radio intrusion of his smartphone emission into an audio device.
Like small cap front the connector barrel to the chassis the immunity they provide is known to be limited at very high frequencies. Recently a member reported radio intrusion of his smartphone emission into an audio device.
The cap solution may be limited, but it is still much better than no connection at all, and is usually enough to prevent problems. Input barrels direct to chassis could also be considered, dare I say, overengineering.
What is your solution to making this connection without causing other problems?
What is your solution to making this connection without causing other problems?
I have'nt since many many years. Maintaining parallel the wires of an unbalanced stereo connection is the rule number one. That's the first recommandation I made it when I begun to particapte to fora in 2002 : a guy complained about hum in his system which had luxorious separated 8 m unbalanced cables between his preamp and his power amps. Making the cables wery close to each other was the cure. Later, I read Bob Cordell made such a recommandation.
Rule number two is to use low impedance sources, less than 100 Ohm is very common now. Using a slightly sophisticated feedback network, ircuit, less than 1 Ohm is acheivable from a single op-amp.
Suggestion : connect all the chassis together with a bar having a very low impedance to approach equipotentiality between them. Multiplying ground loops does not necessarily increase hum.
This is good Audio Component Grounding and Interconnection
From it:
"Loops aren’t bad – it depends on what is on the loop. Unless there is a voltage generator to drive a current around the loop, or radiated current into the loop, it is merely a parallel path. Consider the parallel shields of a left and right channel stereo cable. However, parallel paths do form a loop antenna and can pick up RFI by radiated coupling. Therefore minimize the use of parallel paths to only where necessary and then minimize the area of the loop."
From it:
"Loops aren’t bad – it depends on what is on the loop. Unless there is a voltage generator to drive a current around the loop, or radiated current into the loop, it is merely a parallel path. Consider the parallel shields of a left and right channel stereo cable. However, parallel paths do form a loop antenna and can pick up RFI by radiated coupling. Therefore minimize the use of parallel paths to only where necessary and then minimize the area of the loop."